Repository files navigation

Unofficial Pytorch(1.0+) implementation of nips paperUniversal Style Transfer via Feature Transforms.

Original torch implementation from the author can be foundhere.

Other implementations such asPytorch_implementation1 ,Pytorch_implementation2 orPytorch_implementation3are also available.

This repository provides a pre-trained model for you to generate your own image given content image and style image.

If you have any question, please feel free to contact me. (Language in English/Japanese/Chinese will be ok!)

I propose a structure-emphasized multimodal style transfer(SEMST), feel free to use ithere.

• Python 3.7
• PyTorch 1.0+
• TorchVision
• Pillow

Anaconda environment recommended here!

(optional)

• GPU environment

1. Clone this repository

   git clone https://github.com/irasin/Pytorch_WCTcd Pytorch_WCT

2. Prepare your content image and style image. I provide some in thecontent andstyle and you can try to use them easily.

3. Download the pretrained modelhere and put them under the directory namedmodel_state

4. Generate the output image. A transferred output image and a content_output_pair image and a NST_demo_like image will be generated.

   pythontest.py-ccontent_image_path-sstyle_image_path

   usage: test.py [-h]                [--content CONTENT]                [--style STYLE]               [--output_name OUTPUT_NAME]                [--alpha ALPHA]                [--gpu GPU]               [--model_state_path MODEL_STATE_PATH]

   If output_name is not given, it will use the combination of content image name and style image name.

Some results of content image and my cat (called Sora) will be shown here.